Interpolation Schemes for Collocation Solutions of Two-Point Boundary Value Problems (Q3774879)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 4035049
Language Label Description Also known as
default for all languages
No label defined
    English
    Interpolation Schemes for Collocation Solutions of Two-Point Boundary Value Problems
    scientific article; zbMATH DE number 4035049

      Statements

      Interpolation Schemes for Collocation Solutions of Two-Point Boundary Value Problems (English)
      0 references
      1986
      0 references
      spline interpolation
      0 references
      comparison of methods
      0 references
      numerical examples
      0 references
      interpolatory algorithms
      0 references
      collocation
      0 references
      asymptotic errors
      0 references
      operation counts
      0 references
      storage requirements
      0 references
      superconvergent meshpoint approximations
      0 references
      stability
      0 references
      0 references
      Four interpolatory algorithms for collocation solutions to systems of two point boundary value problems are compared in terms of their asymptotic errors, operation counts and storage requirements. Algorithm 1 computes the collocation solution using a local monomial representation, while the other three all use superconvergent meshpoint approximations and differ only in how the remaining approximations are generated. Algorithm 2 uses the differential equation and either exact derivatives of difference approximations, algorithm 3 uses superconvergent data from nearby intervals, while algorithm 4 is similar to algorithm 3 but a different polynomial is used for each desired derivative approximation. As well as comparing these methods, some new results on the stability of piecewise polynomial interpolants are given. Based on theoretical considerations and numerical testing it is concluded that algorithm 4 is inferior due to problems with highly nonuniform meshes, algorithm 2 is expensive because of extra function evaluations and that algorithm 3 is attractive because is requires less storage and computation and produces answers with accuracy comparable to the collocation solution.
      0 references

      Identifiers

      0 references
      0 references
      0 references
      0 references
      0 references
      0 references